ACL 2026 — Speech & Audio

Speech & Audio

82 papers Links not yet available — ACL proceedings pending on ACL Anthology.

• Dual-Axis Generative Reward Model Toward Semantic and Turn-taking Robustness in Interactive Spoken Dialogue Models
• RespiraMFM: A Multimodal Foundation Model with Contrastive Audio-Language Alignment for Respiratory Disease Identification
• Is EEG-to-Text Feasible in Real-World Scenarios? An In-Depth Analysis Using a Neuropsychology-Inspired Benchmark
• ControlAudio: Tackling Text-Guided, Timing-Indicated and Intelligible Audio Generation via Progressive Diffusion Modeling
• Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech
• When Misinformation Speaks and Converses: Rethinking Fact-Checking in Audio Platforms
• RealTalk-CN: A Realistic Chinese Speech Task-Oriented Dialogue Benchmark with Cross-Modal Analysis
• SAC: Neural Speech Codec with Semantic-Acoustic Dual-Stream Quantization
• FC-TTS: Style and Timbre Control in Zero-Shot Text-to-Speech with Disentangled Speech Representations
• SDiaReward: Modeling and Benchmarking Spoken Dialogue Rewards with Modality and Colloquialness
• Towards Fine-Grained and Multi-Granular Contrastive Language-Speech Pre-training
• Standard-to-Dialect Transfer Trends Differ across Text and Speech: A Case Study on Intent and Topic Classification in German Dialects
• SpeechLLM-as-Judges: Towards General and Interpretable Speech Quality Evaluation
• On the Emotion Understanding of Synthesized Speech
• Beyond Transcripts: A Renewed Perspective on Audio Chaptering
• Shanks: Simultaneous Hearing and Thinking for Spoken Language Models
• Evaluating the Expressive Appropriateness of Speech in Rich Contexts
• UniMoE-Audio: Unified Speech and Music Generation with Dynamic-Capacity Mixture-of-Experts
• XY-Tokenizer: Mitigating the Semantic-Acoustic Conflict in Low-Bitrate Speech Codecs
• VAPO: End-to-end Slide-Enhanced Speech Recognition with Omni-modal Large Language Models
• VoxMind: An End-to-End Agentic Spoken Dialogue System
• An Exploration of Mamba for Speech Self-Supervised Models
• FineLAP: Taming Heterogeneous Supervision for Fine-grained Language-Audio Pretraining
• MeanAudio: Fast and Faithful Text-to-Audio Generation with Mean Flows
• Discovering and Causally Validating Emotion-Sensitive Neurons in Large Audio-Language Models
• Protecting Bystander Privacy via Selective Hearing in Audio LLMs
• Listening Like Humans: Semantics-Guided Noise-Robust Multimodal Speech Recognition
• emg2speech: synthesizing speech from electromyography using self-supervised speech models
• A Data-Centric Approach to Generalizable Speech Deepfake Detection
• POWSM: A Phonetic Open Whisper-Style Speech Foundation Model
• PRiSM: Benchmarking Phone Realization in Speech Models
• Closing the Modality Reasoning Gap for Speech Large Language Models
• DisCo_Speech: Controllable Zero-Shot Speech Generation with A Disentangled Speech Codec
• SEE: Signal Embedding Energy for Quantifying Noise Interference in Large Audio Language Models
• MTAVG-Bench: A Diagnostic Benchmark for Multi-Talker Dialogue-Centric Audio-Video Generation
• SpeakerSleuth: Can Large Audio-Language Models Judge Speaker Consistency across Multi-turn Dialogues?
• Retrieving to Recover: Towards Incomplete Audio-Visual Question Answering via Semantic-consistent Purification
• Rectifying the Emotional Flow: Aligning Priors and Dynamic Guidance for High-Arousal Text-to-Speech
• ATIR: Towards Audio-Text Interleaved Contextual Retrieval
• Omni-Embed-Audio: Leveraging Multimodal LLMs for Robust Audio-Text Retrieval
• PEAP: Proactive Embodied Action Sequence Planning with Joint Understanding of Vision and Audio Perception
• TED-TTS: Training-Free Intra-Utterance Emotion and Duration Control for Text-to-Speech Synthesis
• LLM-ForcedAligner: A Non-Autoregressive and Accurate LLM-Based Forced Aligner for Multilingual and Long-Form Speech
• Mixture-of-Experts with Intermediate CTC Supervision for Accented Speech Recognition
• SAFE-QAQ: End-to-End Slow-Thinking Audio-Text Fraud Detection via Reinforcement Learning
• Audio Jailbreak: An Open Comprehensive Benchmark for Jailbreaking Large Audio-Language Models
• SoundBreak: A Systematic Study of Audio-Only Adversarial Attacks on Trimodal Models
• Towards Fine-grained Audio Captioning with Multimodal Contextual Fusion
• UniSonate: A Unified Model for Speech, Music, and Sound Effect Generation with Text Instructions
• Difference in Task Performance on Sparse Speech Representations
• SpidR-Adapt: A Universal Speech Representation Model for Few-Shot Adaptation
• SpeechMedAssist: Efficiently and Effectively Adapting Speech Language Models for Medical Consultation
• UniVocal: Unified Speech-Singing Code-Switching Synthesis
• From Naturalness to Norms: Interactional Cultural Competence for SpeechLMs
• LLM-Based Multi-Task Bangla Hate Speech Detection: Type, Severity, and Target
• Revisiting Audio-language Pretraining for Learning General-purpose Audio Representation
• Confident, Calibrated, or Complicit: Safety Alignment and Ideological Bias in LLM Hate Speech Detection
• S2S-Arena: Evaluating Paralinguistic Instruction Following in Speech-to-Speech Models
• LCMA-SRT: Language-Conditional Mixture-of-Experts Adapters for Joint Multilingual Speech Recognition and Translation
• Audio MultiChallenge: A Multi-Turn Evaluation of Spoken Dialogue Systems on Natural Human Interaction
• Probing Audio-Visual Reasoning in Multimodal Language Models through the Lens of Audio
• ES4R: Speech Encoding Based on Prepositive Affective Modeling for Empathetic Response Generation
• ImmersiveTTS: Environment-Aware Text-to-Speech with Multimodal Diffusion Transformer and Domain-Specific Representation Alignment
• HalluAudio: A Comprehensive Benchmark for Hallucination Detection in Large Audio-Language Models
• S^4: Operationalizing Speech Act Theory for Strategic Semi-Structured Psychiatric Interview
• Data-efficient Targeted Token-level Preference Optimization for LLM-based Text-to-Speech
• Emotion-Wheel-Guided Audio-Referred Text Representation for Multimodal Emotion Recognition in Conversation
• Still Between Us? Evaluating and Improving Voice Assistant Robustness to Third-Party Interruptions
• Temporal Precision Matters: Brain-Tuning Speech Language Models with Millisecond-Resolution Neural Signals
• CIS-BWE: Chaos-Informed Speech Bandwidth Extension
• TagSpeech: End-to-End Multi-Speaker ASR and Diarization with Fine-Grained Temporal Grounding
• Prosody as Supervision: Bridging the Non-Verbal–Verbal for Multilingual Speech Emotion Recognition
• Measuring User’s Mental Models of Speech Translation in Human-AI Collaboration
• AV-Dialog: Spoken Dialogue Models with Audio-Visual Input
• Speech-Hands: A Self-Reflection Voice Agentic Approach to Speech Recognition and Audio Reasoning with Omni Perception
• PodBench: A Comprehensive Benchmark for Instruction-Aware Audio-Oriented Podcast Script Generation
• BanHADEX: Towards Explainable HAte Speech Detection in Bangla Using Human Annotated EXplanation
• Speculative End-Turn Detector for Efficient Speech Chatbot Assistant
• RV-HATE: Reinforced Multi-Module Voting for Implicit Hate Speech Detection
• Mind the Pause: Disfluency-Aware Objective Tuning for Multilingual Speech Correction with LLMs
• UniSRM: A Unified Speech Reward Model for Reasoning-Based Fine-grained Assessment
• BatonVoice: An Operationalist Framework for Enhancing Controllable Speech Synthesis with Linguistic Intelligence from LLMs